ESTRO 2021 Abstract Book

S1025

ESTRO 2021

Conclusion This study demonstrates that, at a dose level of 45Gy/25#, it is feasible to incorporate spleen constraints while meeting other planning goals for distal OEC. Spleen optimised cases had higher cord, heart and liver doses although differences are small and may not be clinically significant. PBT plans resulted in comparable MSD compared to VMAT. However, there was wider variation in splenic dose for PBT suggesting a greater sensitivity to individual patient anatomy. In most cases (PBT and VMAT) mandatory constraints were met but for most plans, optimal constraints were exceeded. A greater proportion of VMAT plans met all constraints. VMAT may achieve greater spleen sparing in distal OEC compared to PBT. However, a lower integral dose with PBT may still result in lower rates of G4 lymphopenia. MSD of <10Gy is a feasible optimal constraint for clinical use. Prospective studies are required to ascertain the impact on splenic irradiation on G4 lymphopenia rates and survival outcomes. PO-1242 Image-based deep learning model for predicting pathological response in rectal cancer using post-chemoradiotherapy magnetic resonance imaging J. KIM 1 , B. Jang 2 , C. Song 2 , S.H. Jeon 3 , S. Kang 4 , Y.J. Lee 5 1 Seoul National University Bundang Hospital, Department of Radiation Oncology, Seongnam, Korea Republic of; 2 Seoul National University Bundang Hospital, Department of Radiation Oncology , Seongnam, Korea Republic of; 3 Korea Advanced Institute of Science and Technology, Graduate School of Medical Science and Engineering, Daejon, Korea Republic of; 4 Seoul National University Bundang Hospital, Department of Surgery, Seongnam, Korea Republic of; 5 Seoul National University Bundang Hospital , Department of Radiology, Seongnam, Korea Republic of Purpose or Objective To develop an image-based deep learning model for predicting pathological response in rectal cancer using post-chemoradiotherapy magnetic resonance (MR) imaging. Materials and Methods A total of 466 patients with locally advanced rectal cancer who received preoperative chemoradiotherapy followed by surgical resection were enrolled, among whom 113 (24.3%) were allocated to the holdout testing set. Complete response (pCR) was defined as Dworak tumor regression grade (TRG) 4, while good response Digital Poster: Lower GI (colon, rectum, anus)